1. Field of the Invention
The present invention relates generally to a data storage medium and methods and apparatus for recording data on the medium, and, more particularly, to an erasable optical data storage medium and methods and apparatus for writing, erasing, and reading data on the data storage medium.
2. Discussion of Background and Prior Art
Currently, the practical or commercial techniques for recording data are based substantially on magnetic storage technology. In general, the data are stored on magnetic media, such as discs and tapes, on which logic 1 data bits and logic 0 data bits are represented by the magnetization of a medium. For example, one direction of magnetization of a given location or bit storage area of the data storage medium can represent a logic 1, while another direction of magnetization of that bit storage area can represent a logic 0. Each data bit is written on the medium by using a recording head to magnetize the given bit storage area, and each data bit can be erased by writing another bit over the given bit storage area using the magnetic recording head. Each data bit is read by using the recording head to sense the magnetization of the given bit area.
While the magnetic storage technology is commercially successful and advantageous, since about early in the decade of the 1960's, a recording technique known generically as optical recording has been and continues to be considered a very promising alternative for storage data. Optical recording potentially has significant advantages over magnetic recording, including higher data storage density, higher data rates, and longer data archival capabilities. One type of optical recording that has the highest potential is an optical recording apparatus or system which uses, in lieu of the magnetic recording head, a highly focused laser beam as an ultra-fine recording stylus to write and read data at a very high data rate and recording density, and to erase the data. The system includes an erasable optical data storage medium that responds to the laser beam to store the data. For example, the data storage medium responds to the heat generated by the laser beam to erase and write the data, and responds to the light of the laser beam to read the data.
In one optical recording system, to write a data bit, a laser beam is focused on the erasable data storage medium to heat the medium and, thereby, induce a stable transition from one morphological or physical state, e.g., an amorphous state, to another morphological or physical state, to another morphological or physical state, e.g., a crystalline state. The two physical states have different optical properties which are the optical transmittance and optical reflectance properties of the respective states. Therefore, to read the data bit, light from the laser beam, which is at a lower power level than is used for writing, is focused on the data storage medium and will be transmitted or reflected by the medium depending on the physical state of the medium, thereby representing a logic 1 or logic 0. The data bit can be erased by again heating the material with the laser beam at a higher power level to return the medium to its original physical state.
The above-mentioned erasable optical data storage medium is made of semiconductor or chalcogenide materials that, as already stated, change from one state to another when heated. One problem with this data storage medium is that these changes in state are very small or slight, i.e., the amorphous and crystalline states are not substantially distinguishable optically. Therefore, a high signal-to-noise ratio of a reflected laser beam is not obtainable to distinguish between a logic 1 and a logic 0 upon reading the data bit. Another problem with this data storage medium is that the data rate, in particular the writing and erasing speeds, is undesirably very low, e.g., one microsecond. This results from the relatively long time that is required for a given material to undergo the change from one physical state to the other. Another problem is that a relatively high amount of laser energy or power is required to heat the material so that it can transform from one physical state to the other. Yet another problem is that due to this physical state transformation, the data storage medium will fatigue after a relatively few number of erase/write cycles. This fatigue factor will not be competitive with magnetic storage technology, which can achieve a number of erase/write cycles on the order of one million.
U.S. Pat. No. 4,278,734 to Ohta et al, discloses an optical medium in which the material of the medium changes physical state, leading to an increase or decrease in optical density. Ohta et al appears to have solved the data rate problem in that data may be written or erased quickly, e.g. in 50 nsec. Ohta et al also appears to have solved the contrast problem in that the physical states are distinguishable optically. However, one disadvantage of Ohta et al is that the data cannot be erased, bit-by-bit, since a localized bit area location cannot be achieved. Also, the physical change of state or transformation cannot occur on a surface which has anomalies or irregularities in the surface. Furthermore, the materials used for the optical medium are expensive.
U.S. Pat. No. 4,264,986 to Willis, issued Apr. 28, 1981, discloses another type or erasable optical data storage medium. To write a data bit, a laser beam is focused on the medium to induce, by heating, a volumetric expansion of the bit area being heated, thereby creating a small bump or deformation. The presence of the bump represents one logic state, while the absence of the bump represents the other logic state. Upon this heating, the bit area of the data storage medium changes from one physical state, i.e., crystalline, to another physical state, i.e., amorphous, which is the phenomenon that causes the volumetric expansion, thereby creating the small bump which becomes reversibly fixed. A bit is read by focusing on the bit area a laser beam of lower power than is used for writing, and then detecting the amount or scattering of reflected light. If the bump is present, the reflected light will be substantially scattered, so that the intensity of the detected light will be higher. This read/write recording method is attractive, since it provides a good signal-to-noise ratio to distinguish a logic 1 from a logic 0.
One problem with the erasable optical data storage medium of U.S. Pat. No. 4,264,986 is that the laser beam must raise the material to a high temperature, above the melting point, to erase or remove the bump. This has the disadvantage of requiring high-powered lasers. Moreover, this melting of the bump may not leave the surface of the medium smooth, i.e., ripples can form on the surface upon the cooling of the material. Such a smooth surface is needed to be able to continually reproduce a satisfactory bump for properly reading the data bit. Yet another problem is that the data rate is relatively slow due to the need for the change in physical state of the material of the medium, which is a function of relatively slow cooling rates. Furthermore, fatigue, resulting from the physical change of state, is a problem with this data storage medium in that the number of erase/write cycles which can be achieved is only about one thousand.
U.S. Pat. No. 4,371,954 to Cornet, issued Feb. 1, 1983, discloses an erasable optical data storage medium including a substrate, having a low coefficient of thermal expansion, which supports a dual layer having a bottom layer of material and a top layer of material. The bottom layer of material is a relatively inextensible metal or polymer having a high coefficient of thermal expansion, and the top layer is a metal alloy which is in a martensitic phase at ambient temperature and has a low coefficient of thermal expansion. Also, the bottom layer and top layer have a low adhesion to one another, i.e., they are not bonded together, and the latter has a transformation temperature T.sub.t above ambient and below the melting point of the former layer. Above the transformation temperature T.sub.t, the top layer is in its "parent" phase.
To write a data bit, as described in U.S. Pat. No. 4,371,954, a light pulse from a laser beam is absorbed by the dual layer, resulting in the heating of the dual layer at a temperature below the transformation temperature T.sub.t, as well as a differential expansion between the two layers. The bottom layer delaminates or disengages from the substrate and volumetrically expands onto the top layer which forms a bump. Upon cooling, the top layer forms a reversibly fixed bump and the bottom layer contracts back onto the substrate.
To erase the data bit, as described in U.S. Pat. No. 4,371,954, the martensitic top layer is raised to a temperature exceeding the transformation temperature T.sub.t, either by, for example, a higher power laser pulse or a slower displacement or movement of the data storage medium across the laser beam, thereby transforming the top layer to its parent phase. The top layer then contracts onto the bottom layer and, upon cooling, returns to its martensitic phase.
One problem with the erasable optical data storage medium of U.S. Pat. No. 4,371,974 is that the metallic dual layer, and in particular the bottom layer, is relatively inextensible. Consequently, the bump that can be produced is not as high as is desirable for accurately reading the data bit. Another problem is that the top layer must change between the martensitic phase and the parent phase for erasing the data but not for writing the data. One disadvantage of this change of phase is that the erase mode is slow and, concomitantly, cannot occur as quickly as the write mode, thereby requiring significantly different data rates for the respective modes. Another disadvantage is that different laser power pulses are required for writing and erasing the data bit, with the latter being significantly higher, thereby requiring the use of high-powered lasers.
Yet another problem with U.S. Pat. No. 4,371,954 is that the erasable optical data storage medium is highly susceptible to hard bit errors which are errors resulting from imperfections in the medium. More particularly, any anomalies or irregularities in the surface of the medium will affect the ability of the top layer to change between the martensitic and parent phases, resulting in bit errors. Still another problem results from the top layer being metallic or a metal alloy having a low thermal coefficient of expansion. This means that higher power light pulses are needed to expand this type of material, thereby again requiring high-powered lasers. Another problem is that the bottom layer disengages from the substrate upon writing a data bit. This has the disadvantage of enabling the bottom layer to "creep" about the substrate, thereby creating imperfections during use of the medium and preventing the medium from remaining smooth.